Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A planning system for vehicle, the planning system comprising: a plurality of hierarchal software layers including: a mission planning layer comprising one or more neural networks configured to determine an optimal route for the vehicle or mobile robot based on a start point, an end point and a digital map of an environment surrounding the vehicle; a behaviour planning layer comprising one or more neural networks, the behaviour planning layer configured to receive the optimal route determined by the mission planning layer and sensor data sensed by a plurality of sensors of the vehicle, each neural network of the behaviour planning layer configured to predict a respective behavior task for the vehicle based on the sensor data and the optimal route; and a motion planning layer comprising one or more neural networks, the motion planning layer configured to receive each behaviour task predicted by the behaviour planning layer and the sensor data, each neural network of the motion planning layer configured to predict a respective motion task for the vehicle based on the received behavior tasks and the sensor data, wherein the behaviour planning layer is feed-associated with the motion planning layer, and wherein the behaviour planning layer is configured to feed-forward information to the motion planning layer and receive feedback of information from the motion planning layer.
The planning system is designed for autonomous vehicles or mobile robots, addressing the challenge of navigating complex environments while balancing high-level mission objectives with real-time operational constraints. The system employs a hierarchical architecture with three interconnected neural network-based layers. The mission planning layer determines an optimal route from a start point to an end point using a digital map of the surrounding environment. This route is passed to the behavior planning layer, which processes sensor data from the vehicle and predicts specific behavior tasks, such as lane changes or obstacle avoidance, based on both the optimal route and real-time sensor inputs. The motion planning layer then receives these behavior tasks and sensor data to predict precise motion tasks, such as steering angles or acceleration commands, ensuring smooth and safe execution. The behavior and motion planning layers are feed-associated, allowing bidirectional information flow: the behavior layer provides forward guidance to the motion layer, while the motion layer can provide feedback to refine behavior decisions. This hierarchical, feedback-driven approach enables adaptive, real-time navigation while maintaining alignment with the overall mission objectives.
2. The planning system of claim 1 , wherein the mission planning layer is feed-associated with the behaviour planning layer, and wherein the mission planning layer is configured to receive feedback of at least one of information, control, and metadata from the behaviour planning layer.
This invention relates to a planning system for autonomous or semi-autonomous systems, particularly for applications in robotics, autonomous vehicles, or other AI-driven decision-making environments. The system addresses the challenge of integrating high-level mission objectives with real-time adaptive behavior, ensuring that mission goals remain aligned with dynamic operational conditions. The system includes a mission planning layer responsible for defining and managing high-level objectives, such as navigation routes, task priorities, or operational constraints. A behavior planning layer handles real-time decision-making, such as obstacle avoidance, path adjustments, or environmental responses. These layers are feed-associated, meaning the mission planning layer receives feedback from the behavior planning layer. This feedback may include information, control signals, or metadata, allowing the mission planning layer to dynamically adjust mission parameters based on real-time conditions. For example, if the behavior planning layer detects an obstacle, it can relay this information to the mission planning layer, which may then modify the mission route or priorities accordingly. This bidirectional interaction ensures that the system remains responsive to environmental changes while maintaining alignment with overarching mission goals. The system improves adaptability and efficiency in autonomous operations by enabling continuous feedback-driven adjustments.
3. The planning system of claim 1 , wherein each neural network of the behaviour planning layer is feed-associated with each neural network of the a-hierarchically adjacent motion planning layer.
This invention relates to a hierarchical neural network-based planning system for autonomous systems, such as robots or vehicles, that improves coordination between behavior planning and motion planning layers. The system addresses the challenge of integrating high-level decision-making with low-level motion control in a way that ensures smooth, efficient, and context-aware navigation. The core innovation involves a multi-layered architecture where a behavior planning layer generates high-level action plans, and a motion planning layer translates these into executable motion commands. Each neural network in the behavior planning layer is directly connected to every neural network in the adjacent motion planning layer, forming a feed-associated structure. This ensures that motion planning networks receive direct, unfiltered inputs from behavior planning networks, eliminating intermediate processing bottlenecks and improving responsiveness. The system dynamically adjusts planning parameters based on real-time environmental feedback, allowing for adaptive behavior in complex or unpredictable scenarios. The hierarchical design enables modularity, where individual layers can be updated or replaced without disrupting the entire system. This approach enhances scalability and adaptability, making it suitable for applications requiring precise, real-time decision-making in dynamic environments.
4. The planning system of claim 1 wherein the mission planning task comprises one or more neural networks configured to determining one or more checkpoints along the optimal route, and calculating any associated tolls along the optimal route.
This invention relates to an advanced mission planning system designed to optimize routes for autonomous or semi-autonomous vehicles, particularly in complex environments where traditional navigation systems may fail. The system addresses the challenge of efficiently determining optimal paths while accounting for dynamic constraints such as traffic, road conditions, and regulatory requirements. The core innovation involves a neural network-based approach to mission planning, which dynamically generates checkpoints along a route and calculates associated tolls or costs. These checkpoints serve as critical waypoints that ensure the vehicle adheres to the planned trajectory while minimizing deviations. The neural networks are trained to assess real-time data, such as traffic patterns, road closures, and toll pricing, to adjust the route dynamically. This ensures the vehicle follows the most cost-effective and time-efficient path while complying with legal and operational constraints. The system also integrates with other components, such as route optimization algorithms and obstacle avoidance modules, to provide a comprehensive solution for autonomous navigation. By leveraging machine learning, the system improves adaptability and accuracy in mission planning, reducing reliance on pre-programmed routes and enhancing operational flexibility.
5. The planning system of claim 1 wherein the respective behavior task is associated with a respective behavior of the autonomous vehicle.
Autonomous vehicle planning systems are designed to manage the complex decision-making processes required for safe and efficient navigation. A key challenge is ensuring that the system can effectively execute specific behaviors, such as lane changes, turns, or obstacle avoidance, in real-time while adhering to traffic rules and environmental constraints. To address this, a planning system may include a behavior task associated with a specific behavior of the autonomous vehicle. The behavior task defines the actions the vehicle must perform to achieve a desired behavior, such as adjusting speed, steering angle, or braking force. The system may also include a trajectory generator that produces a trajectory based on the behavior task, ensuring the vehicle follows a safe and optimal path. Additionally, the system may incorporate a cost function to evaluate the feasibility and efficiency of the generated trajectory, optimizing for factors like comfort, safety, and adherence to traffic regulations. By associating behavior tasks with specific vehicle behaviors, the planning system can dynamically adapt to changing conditions, improving overall performance and reliability in autonomous driving scenarios.
6. The planning system of claim 5 wherein each respective behavior task comprises one of: changing a lane, waiting at an intersection, passing another vehicle or mobile robot, giving way to the another vehicle or mobile robot, or waiting at an intersection.
Autonomous vehicle and mobile robot navigation systems often struggle with efficiently planning and executing complex behaviors in dynamic environments, such as urban traffic scenarios. Existing systems may lack the ability to dynamically adjust to real-time conditions, leading to inefficient or unsafe maneuvers. This invention addresses these challenges by providing a planning system that generates and executes behavior tasks for autonomous vehicles or mobile robots. The system dynamically selects and prioritizes specific behavior tasks based on real-time environmental conditions and vehicle or robot state. Each behavior task involves a distinct maneuver, such as changing lanes, waiting at an intersection, passing another vehicle or mobile robot, yielding to another vehicle or mobile robot, or waiting at an intersection. The system evaluates multiple possible tasks, selects the most appropriate one, and executes it while continuously monitoring and adjusting the plan to ensure safe and efficient navigation. This approach improves adaptability in complex traffic scenarios, reducing conflicts and enhancing overall system performance.
7. The planning system of claim 1 wherein each respective motion task comprising one of avoiding an obstacle, finding a local path, or controlling the speed, direction or position of the vehicle or the mobile robot.
This invention relates to a planning system for autonomous vehicles or mobile robots, addressing the challenge of efficiently executing motion tasks in dynamic environments. The system generates a motion plan by decomposing it into multiple motion tasks, each assigned to a different processor or processing unit. Each motion task involves either avoiding obstacles, finding a local path, or controlling the vehicle's speed, direction, or position. The system dynamically allocates these tasks to available processors, ensuring parallel processing for improved efficiency. The motion tasks are prioritized based on their importance, and the system adjusts task allocation in real-time to adapt to changing conditions. The planning system also includes a task scheduler that manages the execution sequence of the motion tasks, ensuring that critical tasks are processed first. The system further includes a task monitor that tracks the progress of each motion task and reallocates tasks if a processor fails or becomes overloaded. This approach enhances the system's robustness and responsiveness in navigating complex environments.
8. The planning system of claim 7 wherein the respective motion task is performed by controlling at least one operable element of the vehicle, wherein the at least one operable element includes a GPS unit, steering unit, a brake unit, or a throttle unit.
This invention relates to a planning system for autonomous vehicles, addressing the challenge of efficiently executing motion tasks to navigate and control the vehicle in dynamic environments. The system generates a motion plan for the vehicle by determining a sequence of motion tasks, each involving a specific action such as changing lanes, turning, or adjusting speed. The system then executes these tasks by controlling one or more operable elements of the vehicle, including a GPS unit for location tracking, a steering unit for directional control, a brake unit for deceleration, or a throttle unit for acceleration. The planning system ensures safe and coordinated movement by dynamically adjusting the vehicle's actions based on real-time data, such as sensor inputs or environmental conditions. This approach enhances autonomous driving capabilities by integrating precise control of vehicle components with adaptive planning, improving navigation accuracy and responsiveness in various driving scenarios. The system's ability to manage multiple operable elements allows for seamless execution of complex maneuvers, ensuring efficient and safe operation of the autonomous vehicle.
9. The planning system of claim 1 wherein the sensor data comprises one or more of image data, LIDAR data, RADAR data, global positioning system (GPS) data, and inertial measurement unit (IMU) data, and processed sensor data.
This invention relates to a planning system for autonomous vehicles, addressing the challenge of accurately perceiving and navigating complex environments. The system integrates multiple sensor inputs to generate a comprehensive understanding of the vehicle's surroundings, enabling safe and efficient path planning. The sensor data includes image data from cameras, LIDAR data for precise distance measurements, RADAR data for detecting objects and their velocities, GPS data for location tracking, and IMU data for orientation and motion sensing. Additionally, the system processes this raw sensor data to extract meaningful features, such as object detection, lane markings, and traffic signals, which are critical for decision-making. By fusing these diverse data sources, the system improves situational awareness, reduces uncertainty, and enhances the reliability of autonomous navigation. The processed sensor data is used to construct a detailed environmental model, which the planning system leverages to determine optimal routes, avoid obstacles, and comply with traffic rules. This multi-sensor fusion approach ensures robust performance in varying conditions, such as low visibility or dynamic traffic scenarios, ultimately improving the safety and efficiency of autonomous vehicle operations.
10. The system of claim 1 wherein the sensor data includes data sensed by one or more of a camera, a LIDAR system, a RADAR system, a global position system, and an inertial measurement unit of the vehicle.
This invention relates to a vehicle-based sensor system designed to enhance situational awareness and navigation. The system collects and processes sensor data from multiple sources to improve vehicle operations, particularly in autonomous or semi-autonomous driving scenarios. The primary problem addressed is the need for reliable, multi-modal sensor integration to ensure accurate environmental perception and vehicle positioning. The system incorporates data from various sensors, including cameras, LIDAR (Light Detection and Ranging), RADAR (Radio Detection and Ranging), GPS (Global Positioning System), and an inertial measurement unit (IMU). Cameras provide visual data for object detection and recognition, while LIDAR and RADAR offer precise distance and velocity measurements. GPS delivers global positioning information, and the IMU tracks the vehicle's motion and orientation. By combining these inputs, the system achieves robust environmental mapping, obstacle detection, and navigation capabilities. The integration of these sensors allows the system to compensate for individual sensor limitations, such as camera occlusions or GPS signal interference, ensuring continuous and reliable operation. This multi-sensor fusion approach enhances the vehicle's ability to navigate complex environments, avoid collisions, and maintain accurate positioning under varying conditions. The system is particularly useful in autonomous vehicles, where real-time, high-fidelity sensor data is critical for safe and efficient operation.
11. The planning system of claim 1 wherein each of the one or more neural networks of each of the mission planning layer, the behaviour planning layer, and the motion planning layer are trained offline with new training data before autonomous operation of the vehicle or mobile robot.
This invention relates to an autonomous vehicle or mobile robot planning system that uses multiple neural networks across different planning layers to improve decision-making and navigation. The system addresses the challenge of ensuring reliable and adaptive autonomous operation by leveraging offline training of neural networks before deployment. The planning system includes a mission planning layer, a behavior planning layer, and a motion planning layer, each utilizing one or more neural networks. These neural networks are trained offline with new training data before the vehicle or robot begins autonomous operation. Offline training allows the system to incorporate updated or additional data to enhance performance without requiring real-time learning, which can be computationally expensive or unreliable. The mission planning layer determines high-level objectives, such as route selection or task prioritization. The behavior planning layer translates these objectives into specific actions, like lane changes or obstacle avoidance. The motion planning layer generates precise control commands for vehicle movement, such as steering and acceleration. By pre-training the neural networks offline, the system ensures that the vehicle or robot operates with optimized decision-making capabilities from the start, reducing the risk of errors during autonomous operation. This approach improves efficiency, safety, and adaptability in dynamic environments.
12. The planning system of claim 1 wherein the vehicle comprises a mobile robot, an autonomous vehicle, an autonomous robot, or autonomous drone.
The invention relates to a planning system for autonomous vehicles, including mobile robots, autonomous vehicles, autonomous robots, and autonomous drones. The system addresses the challenge of efficiently navigating and operating in dynamic environments where obstacles, terrain, and mission objectives may change unpredictably. The core planning system generates and evaluates potential paths or action sequences to determine an optimal route or strategy for the vehicle. It incorporates real-time data from sensors and external sources to adapt to environmental changes, ensuring safe and effective navigation. The system may also integrate machine learning or predictive algorithms to improve decision-making over time. By dynamically adjusting to new conditions, the planning system enhances the autonomy and reliability of the vehicle, reducing the need for human intervention. The invention is particularly useful in applications such as logistics, search and rescue, and environmental monitoring, where autonomous vehicles must operate independently in complex and unpredictable settings. The system's adaptability and efficiency make it suitable for various autonomous platforms, from ground-based robots to aerial drones.
13. The planning system of claim 1 , wherein the plurality of hierarchical software layers further comprise a safety planning layer comprising one or more neural networks, each of the one or more neural networks of the safety planning a respective safety task, the respective safety task comprising determining whether a motion planning task corresponding to the respective safety task is safe.
The invention relates to a hierarchical planning system for autonomous systems, particularly for ensuring safe motion planning in dynamic environments. The system addresses the challenge of integrating safety considerations into motion planning to prevent collisions or unsafe operations in autonomous vehicles, robots, or other automated systems. The system includes multiple hierarchical software layers, with a dedicated safety planning layer that evaluates the safety of motion plans generated by other layers. This safety planning layer contains one or more neural networks, each assigned a specific safety task. Each neural network assesses whether a corresponding motion planning task is safe, effectively acting as a safety validator for the motion plan. The safety assessment may involve analyzing environmental data, obstacle detection, trajectory analysis, or other safety-critical factors to determine if the planned motion poses risks. If a motion plan is deemed unsafe, the system may trigger corrective actions, such as recalculating the plan or halting the motion. The hierarchical structure allows the system to separate motion planning from safety validation, ensuring that safety checks are performed independently and robustly. The use of neural networks enables adaptive and data-driven safety assessments, improving reliability in complex or unpredictable environments. This approach enhances the overall safety of autonomous systems by integrating real-time safety evaluations into the planning process.
14. The planning system of claim 1 , wherein the mission planning layer is hierarchically adjacent to the behaviour planning layer and the behaviour planning layer is hierarchically adjacent to the motion planning layer.
This invention relates to a hierarchical planning system for autonomous systems, particularly for robotic or autonomous vehicle navigation. The system addresses the challenge of efficiently coordinating high-level mission objectives with low-level motion control in dynamic environments. The planning system includes multiple layers: a mission planning layer, a behavior planning layer, and a motion planning layer. The mission planning layer generates high-level objectives, such as navigation routes or task sequences, based on mission requirements. The behavior planning layer translates these objectives into intermediate actions, considering environmental constraints and dynamic obstacles. The motion planning layer executes precise motion commands to achieve the desired behavior. The key innovation is the hierarchical adjacency between these layers, ensuring seamless coordination. The mission planning layer directly interfaces with the behavior planning layer, which in turn directly interfaces with the motion planning layer. This adjacency minimizes latency and improves responsiveness by reducing the need for intermediate translation steps. The system dynamically adjusts planning parameters based on real-time feedback, enhancing adaptability in complex environments. The hierarchical structure allows for modular updates, where improvements in one layer do not require overhauls in others. This approach optimizes computational efficiency and reliability in autonomous navigation tasks.
15. The planning system of claim 1 , wherein the one or more neural networks of the mission planning layer are configured to determine the optimal route for the vehicle based on one or more of driving rules, a distance from the start point to the end point and a determination of a shortest distance from the start point to the end point, and a shortest time to travel from the start point to the end point giving the presence of any fixed obstacles between the vehicle and the end point.
A mission planning system for autonomous vehicles uses neural networks to determine optimal routes. The system addresses the challenge of efficiently navigating from a start point to an end point while adhering to driving rules, avoiding obstacles, and optimizing for distance or time. The neural networks in the mission planning layer analyze multiple factors, including driving rules, the direct distance between the start and end points, and the shortest possible path considering fixed obstacles. The system calculates both the shortest distance and the shortest time to reach the destination, accounting for any obstacles that may block the direct path. This ensures the vehicle follows a compliant, efficient route while avoiding collisions or delays. The neural networks are trained to balance these constraints dynamically, adapting to real-time conditions to provide the most effective path. The system integrates these calculations into a broader mission planning framework, enabling autonomous vehicles to navigate complex environments safely and efficiently.
16. A method of controlling autonomous operation of a vehicle or mobile robot, the method comprising: receiving a plurality of behaviour tasks predicted by a behaviour planning layer of a planning system of the vehicle or mobile, each respective behaviour task predicted by a respective neural network of a plurality of neural networks of the behaviour planning layer based on an optimal route determined by a mission planning layer of the planning system and a sensor data received from a sensor mounted to the vehicle or mobile robot; receiving a plurality of motion tasks from a mission planning layer of a planning system of the vehicle or robot, each respective motion task predicted by a respective neural network of a plurality of neural networks of the motion planning layer based on the behaviour tasks received from the behaviour planning layer and the sensor data, wherein the behaviour planning layer is feed-associated with the motion planning layer, and wherein the behaviour planning layer is configured to feed-forward information to the motion planning layer and receive feedback of information from the motion planning layer; and controlling the vehicle or mobile robot based on the predicted behaviour tasks and the predicted motion tasks to operate the vehicle or mobile robot autonomously to navigate the optimal route.
The invention relates to autonomous navigation systems for vehicles or mobile robots, addressing the challenge of integrating behavior and motion planning in a hierarchical, feedback-driven architecture. The system employs a multi-layered planning approach where a mission planning layer determines an optimal route, which is then refined by a behavior planning layer. The behavior planning layer uses multiple neural networks to predict behavior tasks based on sensor data and the optimal route. These behavior tasks are then fed into a motion planning layer, which also uses multiple neural networks to generate motion tasks. The motion planning layer receives feed-forward information from the behavior planning layer and provides feedback to it, creating a dynamic, adaptive planning process. The system controls the vehicle or robot by executing the predicted behavior and motion tasks, enabling autonomous navigation along the optimal route. This architecture improves decision-making by leveraging neural networks at both planning levels and ensuring continuous feedback between layers for real-time adjustments. The invention enhances autonomy by combining high-level route planning with low-level motion control in a unified, feedback-driven framework.
17. A computer program product comprising instructions which, when the program is executed by a computer cause the computer to carry out the method of claim 16 .
A system and method for optimizing data processing in a distributed computing environment addresses inefficiencies in task allocation and resource utilization. The invention involves dynamically assigning computational tasks to nodes within a network based on real-time performance metrics, such as processing speed, memory availability, and network latency. By continuously monitoring these metrics, the system identifies optimal nodes for task execution, reducing bottlenecks and improving overall system throughput. The method includes collecting performance data from each node, analyzing the data to determine task suitability, and redistributing tasks to balance the workload across the network. Additionally, the system may prioritize tasks based on urgency or resource requirements, ensuring critical operations are handled efficiently. The invention also includes a feedback mechanism that adjusts task allocation strategies based on historical performance data, allowing the system to adapt to changing conditions. This approach enhances scalability and reliability in distributed computing environments, particularly in applications requiring high availability and low latency, such as cloud computing, big data analytics, and real-time processing systems. The computer program product implements this method, enabling seamless integration into existing distributed computing frameworks.
18. The method of claim 16 , wherein controlling comprises controlling one or more operable elements of the vehicle or mobile robot to cause the vehicle or mobile robot to perform each behaviour task and each motion task to operate the vehicle or mobile robot autonomously to navigate the optimal route.
This invention relates to autonomous navigation systems for vehicles or mobile robots, addressing the challenge of efficiently navigating an optimal route while performing multiple tasks. The system controls one or more operable elements of the vehicle or mobile robot to execute both behavior tasks (e.g., obstacle avoidance, path planning) and motion tasks (e.g., steering, acceleration) in a coordinated manner. By integrating these tasks, the system enables the vehicle or mobile robot to operate autonomously, dynamically adjusting its movements to follow the most efficient path while avoiding obstacles and adhering to operational constraints. The method ensures seamless coordination between high-level decision-making (behavior tasks) and low-level control (motion tasks), allowing the vehicle or mobile robot to adapt to real-time environmental changes. This approach enhances navigation efficiency, safety, and reliability in autonomous systems.
19. A non-transitory computer readable medium storing instructions executable by at least one processor of a vehicle to cause the vehicle to perform the method of claim 16 .
A system for autonomous vehicle navigation improves path planning by dynamically adjusting route parameters based on real-time environmental and vehicle conditions. The invention addresses challenges in traditional path planning, such as static route calculations that fail to adapt to changing road conditions, traffic, or vehicle performance. The system continuously monitors environmental factors like weather, road surface conditions, and traffic density, as well as vehicle-specific data such as tire pressure, battery levels, and sensor accuracy. Using this data, the system recalculates optimal path parameters, including speed, steering angle, and acceleration, to ensure safe and efficient navigation. The system also integrates predictive analytics to anticipate potential hazards or obstacles, allowing for preemptive adjustments. By dynamically updating path parameters in real-time, the system enhances vehicle safety, reduces energy consumption, and improves overall navigation efficiency. The invention is particularly useful for autonomous vehicles operating in unpredictable environments where static path planning is insufficient. The system may be implemented via software stored on a non-transitory computer-readable medium, executable by a vehicle's onboard processor to continuously refine navigation decisions based on real-time data inputs.
Unknown
October 6, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.